The effect of acute migraine attack and rectally given metoclopramide on the absorption of orally given tolfenamic acid (300 mg) was investigated in seven female patients in a crossover study consisting of four phases, two without migraine and two during migraine. Metoclopramide hydrochloride (20 mg) or placebo was given double-blind. Migraine attacks delayed the absorption of tolfenamic acid. Serum concentrations of tolfenamic acid 1.5 and 2 h after drug administration remained smaller, the peak serum concentration (tmax) occurred later and the area under the serum concentration-time curve between zero and 2 h (AUC0-2 h) remained decreased during migraine. Metoclopramide pretreatment in migraine attacks increased the serum concentration of tolfenamic acid at 1.5 h, but its peak concentration, time to peak concentration and the AUC0-5 h remained unchanged as compared with the values obtained with tolfenamic acid alone. Between the absorption of tolfenamic acid without migraine and after metoclopramide pretreatment during migraine no significant differences existed. When the patients were studied without migraine the serum concentrations of tolfenamic acid 45 min and 60 min after its administration were higher after metoclopramide than after placebo pretreatment. During migraine attacks the serum concentrations and the AUC0-5.5 h of metoclopramide were slightly lowered. The impairment of drug absorption by migraine was not related to the duration or severity of the attack. The observed changes in drug absorption during migraine attacks are obviously due to the delay in gastric emptying. Rectally administered metoclopramide accelerates the absorption of orally given tolfenamic acid.